ESXi 5.0 Hardware Status shown as Warning - but no all Disks and SA Battery are fine

[Bernard Burette]

Hi,

Had the same problem, looks like it is linked to buggy HP software, no hardware problem here.

I solved it by installing the previous SMX and AMS from HP (not the ones that are packaged with the ESXi 5.1 custom image for ProLiant), here are the steps :

- download HP ESXi Offline Bundle for VMware ESXi 5.x version 1.3.5 (the buggy version is 1.4) ;

- the link for me was : http://h20000.www2.hp.com/bizsupport/TechSupport/SoftwareDescription.jsp?lang=en&cc=fi&prodTypeId=15351&prodSeriesId=4091412&swItem=MTX-0f7de8802be4423db38c8cb200&prodNameId=4091432&swEnvOID=4115&swLang=8&taskId=135&mode=4&idx=1

- copy the ZIP to a datastore that can be accessed by the ESXi server showing the problem ;

- activate SSH on that ESXi server and log in directly (alternatively you can activate ESXi shell from the console then Alt-F1 and log in directly to the console) ;

- remove the buggy SMX package ;

- the command is : esxcli software vib remove -n hp-smx-provider

- the output of the removal should look like :

  Removal Result
   Message: The update completed successfully, but the system needs to be rebooted for the changes to be effective.
   Reboot Required: true
   VIBs Installed:
   VIBs Removed: Hewlett-Packard_bootbank_hp-smx-provider_500.03.02.00.23-434156
   VIBs Skipped:

- now install the good version from the ZIP ;

- the command is : esxcli software vib install -d /<<full-path>>/hp-esxi5.0uX-bundle-1.3.5-3.zip

- you have to use the full path to the ZIP, a relative path (like ./hp-esxi5.0uX-bundle-1.3.5-3.zip does not work) ;

- since the ZIP is on a datastore the full path will probably look like /vmfs/volumes/<<datastore-name>>/...

- the output of the installation should look like :

  Installation Result
   Message: The update completed successfully, but the system needs to be rebooted for the changes to be effective.
   Reboot Required: true
   VIBs Installed: Hewlett-Packard_bootbank_hp-ams_500.9.2.0-11.434156, Hewlett-Packard_bootbank_hp-smx-provider_500.03.01.10.2-434156
   VIBs Removed: Hewlett-Packard_bootbank_hp-ams_500.9.3.0-13.434156
   VIBs Skipped: Hewlett-Packard_bootbank_char-hpcru_5.0.3.09-1OEM.500.0.0.434156, Hewlett-Packard_bootbank_char-hpilo_500.9.0.0.9-1OEM.500.0.0.434156

- you can see that the AMS part was removed and an older version was installed as well ;

- now you have to *reboot* the ESXi server (just stopping and restarting the agents is not enough) ;

- after the reboot in the vSphere client select you host server and go "hardware status" you can confirm the warning is gone and you can browse through software to see which version of SMX and AMS parts are present.

This solved my issue, hope it helps you folks.

Regards.

[Mike O.]

Other than the HPSA warning in VMWare's hardware status, is there any issue with leaving the SMX/AMS on the version bundled with the latest 5.1 image?

I'd like to take advantage of the other upgraded drivers in the newer image, but instead of going through the upgrade to the latest 5.1 image, then manually downgrading a couple of modules on every blade, could I just leave the "buggy" SMX/AMS versions on there and live with the warning until HP releases a fix?

I'm just wondering if leaving it there will cause any other any issues, or hide any real warning with the disks or battery.

[MercuryGate]

We had the same issue with the hardware status tab consistently showing a warning (reporting a value of 2) for the raid controller.

We had HP run their HP Smart Start program on the raid controller and they determined that there were no issues with the raid controller.  We decided to try to use a previous version of the HP Offline Bundle instead of the latest and greatest:

latest version from HP:  hp-esxi5.0uX-bundle-1.4-16.zip

previous version:  hp-esxi5.0uX-bundle-1.3.5-3.zip

When using the older version the issue no longer appears (hp-esxi5.0uX-bundle-1.3.5-3.zip).  So it appears that this is an issue with the latest offline bundle

[oaksboard]

Same issue here on a series of DL580 g7 and DL580 g6 servers we recently upgraded to the latest bundle.

Would be nice to know from HP if there are any other issues, besides the warning.

My biggest concern at the moment is that a real warning could remain unnoticed.

One of the servers has an additional problem: after a few days uptime the hardware monitor switches to "unknown" state.

[theweave]

So I have this same thing......I can cleary read the fix but am more interested in the answer, which up to now, seems to have been avoided. 

What is the harm in just leaving this alone until HP puts out an update.  Backdating these to clear a GUI icon is not worth the effort where I sit???

Also, when is HP going to ackowledge this and give an ETA on a real FIX, not a workaround?

-weave

[Stan Gobien]

Same problem here on a DL380p Gen8.

Also see the topic at vmware communities:

http://communities.vmware.com/message/2222964#2222964

I'm waiting for a newer vib that fixes this bug.

[jo_strasser]

Hi,

same here based on ESXi 5.1-latest and ESXi 5.0-latest (BL465c G7 and BL680c G5)

HP: we need a very quick fix ft that pls...

Johannes

[ragmon]

I'm told that the latest HP Offline Bundle fixes the issues with the hardware status warning http://vibsdepot.hp.com/hpq/apr2013/

[didwenger]

On a DL380p G8 with ESXi 5.0 U2, I have installed the latest HP bundle from April 2013 (hp-esxi5.0uX-bundle-1.4.5-3.zip)  and the error is gone. It's also confirmed in this post: http://www.jonmunday.net/2013/04/16/fix-hp-esxi-5-x-management-bundles/ 

[gsedam]

Perfect - Thank you for the detailed explanation.  Works perfect....